Transcription factors play critical roles in many types of cancer but, for a long time, were considered undruggable because of the lack of specific active sites that can be targeted by the types of small molecules that typify most current drugs. However, the use of short peptides has recently emerged as a promising strategy to target transcription factors that function through specific protein complexes, since distinct interaction patterns may confer a high degree of selectivity on the peptide. In this proposal, we seek to use this new strategy to target the leukemogenic fusion protein/transcription factor, AML1-ETO, that is most frequently involved in acute myeloid leukemia. We have found that, in leukemic cells, AML1- ETO resides in a stable protein complex containing multiple transcription factors and cofactors. Within this complex, the dimerized AML1-ETO directly interacts with a family of conventional transcriptional activators, E proteins, that are implicated in hematopoietic lineage developmental events. We also have found that the AML1-ETO dimerization domain (NHR2), which previously was shown to be critical for leukemogenesis, utilizes a distinct surface of the dimerized alpha-helixes to mediate the AML1-ETO interaction with a conserved motif in E proteins. This particular interaction pattern ideally allows the design of inhibitors to specifically disrupt the interaction and to manipulate the activities of AML1-ETO, thus providing a potential target for leukemia treatment. In this regard, and in further support of this NHR2-E protein interaction as a therapeutic target, a specific mutation that disrupts the NHR2-E protein interaction, but not NHR2 dimerization, has been shown to impair the capacity of AML1-ETO to enhance human hematopoietic stem cell self-renewal. Based on these biochemical and functional studies of the AML1-ETO-containing transcription factor/cofactor (AETFC) complex(es), we plan (i) to further identify and characterize the AETFC complex(es) by detailed mechanistic studies; (ii) to identify direct AML1- ETO target genes by genome-wide ChIP analyses; (iii) to clarify, through cell-based and cell-free in vitro transcription systems, the detailed mechanisms by which AML1-ETO and other components cooperate to (de)regulate transcription; (iv) to study (and validate) the biological functions of individual components and their interactions in leukemic cellular and mouse models; and (v) to design specific peptidomimetic inhibitors to manipulate the action of AML1-ETO in transcription and leukemogenesis.